Report to NIH Council on the Overall Progress of the Protein Structure Initiative (PSI)


Start Date: 3/27/2003 7:00 AM

End Date: 3/28/2003 4:00 PM

During the period January-March 2003 members of the Protein Structure Initiative Advisory Committee (PSIAC) visited all nine of the centers supported under the Protein Structure Initiative (PSI). Typically the visiting group included four or five members of the Advisory Committee plus three or four staff members from the NIH. The site visits were followed by a two-day meeting at the NIH, March 27-28, 2003, attended by all members of the PSIAC. During this meeting the overall progress was discussed and plans were considered for the proposed second phase of the initiative.

Overall, members of the Advisory Committee remain very impressed and highly encouraged by the progress that has been made. While none of the centers has reached what could be described as a fully-developed production mode, allowing 100 or more structure determinations per year, very substantial progress by several of the groups has already been made toward this end. In some cases well-planned, highly efficient pipelines have been developed for target selection, cloning and expression, as well as protein purification, crystallization, and structure determination. Several of the consortia reported the successful determination of 50 or so structures with the expectation that these totals should increase substantially during the remaining 2-1/2 years of the pilot phase.

It was anticipated from the outset that the initiative would provide examples of proteins with new folds. It was also anticipated that by determining the structure of one representative from a large family the structures of remaining members of the family could be inferred by homology modeling. It was also anticipated that such studies could provide new and perhaps unexpected functional insights. All of these objectives are beginning to be met and there is every reason to expect that the impact in these areas will increase substantially within the next year or two. As well as this "structure-based" information, the initiative has also led to novel technologies and approaches that could have a large impact on structural biology in general. Examples include the use of nanotechnology to permit detailed structural studies with amounts of material far less than in the past. Also, systematic analysis of large-scale crystallization screens suggests that most successful crystallizations come from a relatively limited subset of conditions. "Data mining" of this sort should improve overall efficiency and benefit the community as a whole. Highly innovative approaches are also being developed to obtain variants of proteins that fold better and/or are more soluble than their native counterparts.

At the same time, some broad problem areas clearly remain. For example, structural studies of membrane proteins remain difficult and challenging. Likewise, proteins that exist in complexes may require special techniques to identify and study. Some of the centers are beginning to address these problem areas.

Members of the PSIAC are especially gratified by the enthusiasm displayed by each of the centers. Certainly, the objective of being able to determine 200 or so new protein structures a year remains a formidable challenge, especially if this total is to include membrane proteins and complexes. Nevertheless, the progress to date is most encouraging and further progress will clearly not be for want of effort.

For the committee: Brian W. Matthews (Chair) April 15, 2003